基于遗传粒子群算法的FBG传感网重叠光谱的解调研究

Demodulation of Overlapping Spectra of FBG Sensor Network Based on Genetic Particle Swarm Optimization

由于光源带宽有一定的范围,当光纤布拉格光栅(FBG)传感器的数量使用过多时会发生光谱重叠现象,所以会造成反射光谱的中心波长识别困难。为解决光谱重叠问题,通过利用光谱形状复用技术构造重叠光谱,并根据遗传算法和粒子群算法的特点,将二者进行结合形成了遗传粒子群混合算法,用于改进识别反射光谱的中心波长的收敛速度和精度。通过仿真实验结果可知,当FBG传感器的反射光谱发生重叠时,遗传粒子群混合算法可以实现对重叠光谱中心波长的识别,且识别误差范围在5 pm之内。该方法为解决因光源带宽而限制FBG传感器的使用数量提供了一种可行方案。

Due to the bandwidth of the light source,with the excessive use of the fiber Bragg grating(FBG)sensor the range of spectral overlap will occur.It is difficult to identify the center wavelength of the reflection spectrum.In order to solve the problem of spectral overlap,constructing overlapping spectra by utilizing spectral shape multiplexing techniques,according to the characteristics of genetic algorithm and particle swarm optimization,a hybrid genetic particle swarm optimization algorithm is formed,which can improve the convergence speed and accuracy of the original algorithm and apply it to solving the problem of spectral overlap.The simulation results show that when the reflection spectrum of FBG sensor overlaps,the hybrid genetic particle swarm optimization algorithm can recognize the central wavelength of overlap spectrum,with the recognition error±5 pm.This method provides a feasible solution to limit the number of FBG sensors due to the bandwidth of the light source.